- 🚀 Founder building AI-powered career-tech products (job discovery + personalization)
- ☁️ I design and deploy GCP-native backends (Cloud Run, Firestore, Storage, BigQuery)
- 🕸️ I build reliable scraping + ingestion pipelines (Playwright, normalization, dedupe)
- 🤖 I work on GenAI extraction pipelines + evaluation (Gemini / Vertex AI)
- 🎯 Interests: Data Engineering, Search/Ranking, Personalization, MLOps
- Built an AI-powered job platform focused on helping job seekers discover relevant roles faster
- Worked on filtering, personalization, and scalable backend systems
- 🌐 Website: https://foxhunt.ai
- Designed scraping pipelines with pagination/scroll handling, deduplication, and recovery logging
- Automated structured metadata extraction and normalization for search-ready storage
- Built resume scoring + suggestions pipeline using Gemini/Vertex-style evaluation
- Focus on structured JSON outputs, actionable feedback, and history tracking
- Production-style APIs for text extraction, job filtering/search, analytics logging, and caching
- Designed for reliability, cost-awareness, and scale
- Languages: Python, SQL, JavaScript/TypeScript, R
- Frameworks: FastAPI
- Cloud: Google Cloud (Cloud Run, Cloud Storage, Firestore, Cloud Functions, IAM, Logging)
- Databases: MongoDB, Firestore, BigQuery
- Automation & DevOps: Docker, GitHub Actions, Linux
- Scraping: Playwright, BeautifulSoup
- AI/ML: Gemini / Vertex AI, NLP, ML/DL
- Processed and analyzed millions of NYC taxi trip records
- Built ETL workflows and ran analytical queries for business insights
- Worked with large-scale tabular data for credit risk modeling
- Trained ML models (including gradient boosting / neural nets) and evaluated performance tradeoffs
- Designed APIs and pipelines with validation, normalization, and structured outputs
- Focused on scalability, monitoring, and clean data contracts
- LinkedIn: https://www.linkedin.com/in/akvkumar/
- Portfolio: https://karthikadari.com/